A

http://dx.doi.org/10.5935/0103-5053.20140036
J. Braz. Chem. Soc., Vol. 25, No. 4, 759-769, 2014.

Article
Printed in Brazil - ©2014 Sociedade Brasileira de Química
0103 - 5053 $6.00+0.00

Study of Biodiesel Photodegradation Through Reactions

Catalyzed by Fenton’s Reagent

Elizangela Ambrosio, Lorena M. Milano, Maísa Tatiane F. de Souza,

Lucas U. R. Chiavelli, Paula F. Montanher, Jesuí V. Visentainer,
Vitor de C. Almeida, Nilson E. Souza* and Juliana C. Garcia

Departamento de Química, Universidade Estadual de Maringá,

Av. Colombo, 5790, 87020-900 Maringá-PR, Brazil

A fotodegradação de biodiesel em contato com água, utilizando reação foto-Fenton, foi

investigada neste estudo. Após 360 h de fotodegradação obteve-se uma redução de 73% dos ésteres
metílicos de ácidos graxos (FAMEs) quantificados inicialmente por cromatografia em fase gasosa
acoplada ao detector por ionização em chama (GC/FID). No decorrer da fotodegradação foram
detectados picos para cetona e grupo epóxido nas cadeias alifáticas em análises por cromatografia
em fase gasosa acoplada ao espectrômetro de massas (GC/MS), e deslocamentos típicos de aldeídos
e ácidos graxos de cadeia curta em análises por espectroscopia de ressonância magnética nuclear de
hidrogênio (1H RMN). As análises de eco toxicidade usando como organismo teste Artemia salina
evidenciaram componentes tóxicos na fase aquosa que aumentaram até 168 h de fotodegradação,
apresentando decréscimo após este período.

This study reports on the photodegradation of biodiesel in contact with water using the
photo‑Fenton reaction. After 360 h of photodegradation, we observed a reduction of 73% in the
amount of fatty acid methyl esters (FAMEs) initially quantified by gas chromatography coupled with
a flame ionization detector (GC/FID). During the photodegradation, peaks for ketones and epoxy
groups in carbon chains were recorded by gas chromatography coupled with mass spectrometry
(GC/MS), and typical aldehyde and short-chain fatty acid shifts in hydrogen nuclear magnetic
resonance spectroscopy (1H NMR) were observed. Ecotoxicity assays with Artemia salina revealed
the presence of toxic components in the aqueous phase in increasing amounts up to 168 h of
photodegradation and decreasing thereafter.

Keywords: photodegradation, Fenton reactions, biodiesel, bioassay, advanced oxidation

processes

Introduction on the oxidative stability of biodiesel in contact with air,

Biodiesel and bio-oil are useful options for the
generation of renewable energy from biomass. Indeed,
aiming at economic and social benefits, biodiesel is already
being used in several countries, including China, Nicaragua,
the United States, Japan, European Union countries,
and Brazil. However, during the transport of these fuels,
accidents and spills may occur and contaminate soil and
aquatic environments.1 Due to limitations in the removal of
this type of material from the environment, it is important
to investigate alternatives for safely and economically
degrading them on site. To date, studies have been published
*e-mail: nesouza@uem.br
light, and at high temperatures.2-6 It is known that the stability
of biodiesel is related to the presence of unsaturation in the
carbon chains, temperature, presence of light, enzymes,
and microorganisms, and storage conditions. Oxidation
can be accelerated in the presence of metallic ions,
leading to the formation of several degradation products,
such as peroxides and hydroperoxides, which can later be
converted into aldehydes, ketones, and short-chain and
volatile carboxylic acids. However, long-chain compounds
are also formed by oxidative polymerization.6 Physico-
chemical processes, such as coagulation/flocculation7,8 and
filtration,9,10 may be used to treat effluents by removing
organic matter. These processes can be complemented
with advanced oxidative processes (AOPs), as performed
760 Study of Biodiesel Photodegradation Through Reactions Catalyzed by Fenton’s Reagent
by Ramirez et al.,11 whereby a biological process was
linked to the photo‑Fenton process to treat wastewater
from the washing of biodiesel. AOPs are efficient for
the degradation of many organic compounds and are
characterized by the action of the highly oxidative hydroxyl
radical (•OH) (equation 1), which can oxidize various
organic compounds to CO2, H2O, and inorganic ions derived
from heteroatoms.12
HO• + e– + H+ → H2O, Eº = 2.730 V (1)
Hydroxyl radicals are generally formed in reactions
as a result of the combination of oxidants (e.g., ozone or
hydrogen peroxide) and catalysts (e.g., metallic ions and
semiconductors) under ultraviolet (UV) or visible (Vis)
light. Among AOPs, the Fenton process is often employed
due to its efficiency and non-toxicity.12-14 The Fenton
reaction occurs through the catalyzed decomposition of
hydrogen peroxide by ferrous ions in an acid medium,
with the production of hydroxyl radicals (equation 2). In
the absence of a substrate, the hydroxyl radicals formed
may oxidize other Fe+2 ions, as shown in equation 3, and
the resulting ferric ions can decompose H2O2 into water and
oxygen. However, excess H2O2 can hinder the degradation
process due to the formation of hydroperoxyl radical (HO2•)
via a reaction with hydroxyl.12
Fe2+ + H2O2 → Fe3+ + HO• + OH– (2)
Fe2+ + HO• → Fe3+ + OH– (3)
The objective of this study was to investigate the
photodegradation of biodiesel B 100 using the photo‑Fenton
AOP under irradiation from a mercury lamp and to compare
it to other catalytic degradation methods (Fenton and
photolysis) in terms of degradation time and efficiency.
Experimental
This study utilized pure biodiesel (B 100) obtained
by a methylic route from animal (40%) and vegetable
origin (60%) and was kindly provided by a mill from the
northwest region of Paraná State, Brazil. The biodiesel
was stored at room temperature in amber-colored
glass flasks wrapped in aluminum foil. The following
chemicals of analytical grade were used without previous
purification: H2O2 (30‑32%, FMaia), H2SO4 (95-97%,
Biotec), FeSO4.7H2O (97.6%, FMaia), ethanol (99.3%,
FMaia), NaOH (98.0%, Synth), K 2 Cr 2 O 7 (99.0%,
Nuclear), KI (99.0%, Neon), Na2SO3 (98.8%, J. T. Baker),
HPLC‑grade dichloromethane (99.9%, Sigma-Aldrich),
and HPLC-grade hexane (99.9%, Mallinckrodt).
J. Braz. Chem. Soc.
Chamber for sample irradiation
A wooden box measuring 0.57 × 0.90 × 0.53 m
(l × w × h) was fitted with 8 cooling blowers, four on each
of two opposite sides. A mercury lamp (250 W, Philips E-40)
without a glass protection cover was fixed at the center of the
top of the box, 30 cm away from the suspension surface. The
lamp emitted in the UVA, UVB, UVC, and visible bands,
but the UVC radiation was filtered by the borosilicate glass
of the reactors. The inside of the box was painted black with
Brasilux extra-quick-drying enamel paint to prevent light
reflection on the walls. The power of the light source inside
the photo-reactor was measured with a light meter probe
placed at the same position of the sample, resulting in an
irradiance of approximately 3.0 mW cm-2.
Biodiesel degradation
To accelerate the degradation of biodiesel, assays
were performed with aliquots of biodiesel in contact with
water and Fenton’s reagent. The water was acidified with
sulfuric acid (pH 2.50) and was placed in Petri dishes
(15.0 ± 0.1 mL), and B 100 was added (1.00 ± 0.02 mL),
forming a thin layer on top of the water. Fenton’s catalyst,
consisting of a ferrous sulfate solution (9.79 ± 0.49 g L-1)
mixed with 30% (v/v) hydrogen peroxide, was then added;
the dishes were covered with PVC film and placed inside
the irradiation chamber.
Temperature effect
To determine the effect of the temperature inside the
reactor on the degradation of biodiesel, some Petri dishes
were wrapped in aluminum foil and placed inside the box
together with the other samples.
The influence of UV radiation
The effect of light on the degradation of biodiesel was
assessed by photolysis assays. Aliquots of B 100 without
Fenton’s reagent were irradiated together with the other
samples.
Initial H2O2 concentration
To determine the optimum degradation conditions, the
irradiation time was set at 144 h, and the ratio between the
molar concentration of ferrous ions [Fe2+] from the ferrous
sulfate heptahydrate solution and the molar concentration
of hydrogen peroxide [H2O2] was varied as follows:
[Fe2+]:[H2O2] = 0.65, 0.30, and 1.33, varying [Fe2+] from
Vol. 25, No. 4, 2014 Ambrosio et al.
5.34 to 10.4 × 10-3 mol L-1 and [H2O2] as 4.02, 8.05, 15.6,
and 32.9 × 10-3 mol L-1.
Extraction
After the irradiated samples were removed from the
chamber, the pH of the samples was measured and adjusted
to 6.5-7.0 by adding a Fenton reaction inhibitor solution
containing 0.10 mol L-1 KI, 0.10 mol L-1 Na2SO3, and
0.10 mol L-1 NaOH to precipitate solid Fe(OH)3.13 The
mixture was filtered through a membrane (HA cellulose
esters, 0.45 µm, 47 mm, flat white, Millipore) using
hexane to facilitate the passage of the organic phase, and
a separation funnel was used to separate the organic and
aqueous phases of the filtrate (hexane, 2  × 10.0 mL). The
organic phase was evaporated in a rotary evaporator and
transferred to Eppendorf tubes, covered with aluminum
foil, and frozen. The volume of the aqueous fractions was
adjusted to 99.919 ± 0.0218 mL and stored in a refrigerator
in amber-colored glass flasks wrapped with aluminum foil.
Gas chromatography coupled with mass spectrometry
(GC/MS)
The chemical constituents were analyzed by GC/MS
(Thermo-Finnigan) equipped with a source of electron
ionization (EI, 70 eV) and a mass analyzer quadrupole.
The constituents were identified by a comparative analysis
with fatty acid methyl ester mass spectra from the spectrum
database contained in the NIST MS Search spectral library
using the software Xcalibur with the Kovats indices, which
were obtained by co-injection of the organic phase with a
mixture of alkane standards (C9-C21) using the Van den
Dool and Kratz equation.15 The oil phases were subjected to
rotary evaporation to eliminate the hexane and diluted with
HPLC-grade dichloromethane at 1:10000 and analyzed by
GC/MS using a DB-5 capillary column (5% phenyl-95%
methylpolysiloxane, 30 m × 0.25 mm × 0.25 µm) and He
as the carrier gas. The detected masses ranged from 45 to
550 m/z. A sample volume of 1 µL was injected in split
mode 1/10 at injector and ionizer temperatures of 250 °C,
with the following heating ramp: initial temperature of
50 °C up to 290 °C at 3 °C min-1, maintained for 20 min.
Gas chromatography coupled with flame ionization detection
(GC/FID)
The methyl esters were separated by gas chromatography
using a Trace Ultra 3300 chromatograph (Thermo Scientific)
equipped with flame ionization and a cyanopropyl capillary
column (100 m × 0.25 mm × 0.25 μm, CP 7420). The
761
gas flow rates used were 1.2 mL min-1 carrier gas (H2),
30 mL min-1 make-up gas (N2), and 35 and 350 mL min-1
flame gases (H2 and synthetic air, respectively). The sample-
splitting rate was 1:80, and 2-µL samples were injected
in triplicate. The operation parameters were as follows:
detector and injection port temperatures of 240 °C and
220 °C, respectively, and a column temperature of 185 °C
for 7.5 min, programmed to increase at 4 °C min-1 from 1 to
235 °C and kept at this temperature for 1.5 min. The peak
areas were determined using ChromQuest 5.0 software.
The fatty acid retention times were compared to those of
standard methyl esters. Quantification (in mg FAME g-1
of biodiesel) was performed using tricosanoic acid methyl
ester as an internal standard (23:0), as described by Joseph
& Ackman.16 Theoretical FID correction factor values were
used to obtain the concentrations,17 and the FAME contents
were calculated in mg g -1 of biodiesel using equation 4:
FAME = (AX WIS CFX /AIS WX ) × 100 (4)
where FAME is the mg of fatty acid methyl esters per g of
biodiesel, AX is the peak area (FAME), AIS is the peak area
of the internal standard (IS), methyl ester of tricosanoic
acid (23:0), WIS is the amount of IS added to the sample
in mg, WX is the sample weight (in mg), CFX is the
theoretical correction factor, and CFAE is the conversion
factor necessary to express the results in mg of fatty acids
rather than methyl esters. However, the CFAE factor was
not used in this study because the results were already
expressed in mg FAME.
Nuclear magnetic resonance spectroscopy (1H NMR)
B 100 biodiesel and biodiesel samples photodegraded
with Fenton’s reagent for 6, 12, 24, 48, 72, 96, and 120 h
were analyzed by 1H NMR.3,18 The spectra were recording
using a Varian Mercury Plus Model BB 300 MHz
spectrometer operating at 300.06 MHz at 26.0 °C. The
samples were dissolved in deuterated dichloromethane
(CDCl3, from Aldrich) at 0.1% v/v tetramethylsilane (TMS)
(δ = 0.0 ppm), and the chemical shifts (δ) are given in ppm
relative to TMS. The following acquisition parameters were
used: spectral window, –2 to 14 ppm; pulse, 45°; number
of repetitions of the pulse sequence, 32; acquisition time,
3.333 s; recycling delay, 1.000 s; spectral width, 4800.8 Hz;
line broadening, 0.2 Hz; standard pulse sequence for 1H.
Oxitest
The biodiesel induction time was assessed using an
Oxitest apparatus with approximately 5.0 g of sample
762 Study of Biodiesel Photodegradation Through Reactions Catalyzed by Fenton’s Reagent J. Braz. Chem. Soc.

per analysis, an initial pressure of 6.0 bar, and an oxygen a Varian Cary-50 commercial fluorimeter using a quartz
temperature of 110 °C. cell. The wavelength interval, Δλ, between λem and λex was
20 nm, with a 10 nm bandpass. All the experiments were
Chemical oxygen demand (COD) performed at 20 °C.23,24

The chemical oxygen demand (COD) of the aqueous UV-Vis spectrophotometry

phase was determined by the colorimetric method.19 To
verify the consumption of the hydrogen peroxide added
at the beginning of the photocatalysis process, it was
necessary to quantify the final amount and its evolution
during the photodegradation process. The ammonium
metavanadate method was used,14,20 and the values of
residual peroxide were subtracted from the COD values.
Toxicological assay with A. salina L.
The aqueous and organic phases were tested using a
Lambda 25 double-beam spectrophotometer (Perkin Elmer)
in the range of 800 to 200 nm with 1.0 cm optical path
length quartz cuvettes. Distilled water was used to adjust the
transmittance to 100%, and aqueous-phase aliquots diluted
in distilled water (1:1000) were analyzed. The organic
phase was diluted in ethanol (1:10000), and ethanol was
used to adjust the transmittance to 100%.

A brine shrimp (A. salina L.) assay was also employed
for lethality screening of the aqueous phase of the samples
irradiated with Fenton’s reagent for 24, 72, 120, 168, and
360 h based on the LC50 criterion using a nutritive solution,
as described by Meyer et al..21,22 The cyst-like eggs were
hatched within a few hours, and the most resistant nauplii
were used in the toxicity assay. For each test, a 5 mL
mixture of effluent sample and nutritive solution (v/v) was
prepared in a 10 mL glass tube at four dilutions (1, 10, 20,
and 50%) in triplicate and was added to 1.00 mL of brine
shrimp mixture into each tube. After 24 h, the numbers of
dead and live brine shrimp in each tube were counted. As
a control assay, A. salina L. larvae were incubated in pure
nutritive solution.
Fluorescence spectroscopy
Synchronous fluorescence scans were recorded for the
organic phase irradiated for 24, 72, 120, 168, and 360 h
with Fenton’s reagent and for the initial biodiesel B 100
sample without dilution. The spectra were obtained with
Results and Discussion
Determination of the best biodiesel photodegradation
conditions using Fenton’s reagent
Eight biodiesel photodegradation assays were performed
with five repetitions using a fixed degradation time of 144 h,
Fenton reagent volumes ranging from 1.5 to 3.0 mL, and
ferrous ion to hydrogen peroxide rates [Fe2+]:[H2O2] of
0.30, 0.65, and 1.33, as shown in Table 1.
COD analyses of the aqueous phase and GC/FID
analyses of the organic phase were performed, and the
results are shown in Table 2.
Table 2 also shows the results of the main FAMEs
identified in B 100, which are expressed as the methyl
esters identified in 1 g of biodiesel or the organic phase.
Comparing the FAME values for the initial B 100
biodiesel with the values found in experiments 1, 4, and
7 revealed an increase in the concentration of FAMEs
in these experiments (Table 2). These results can be
explained by the initial B 100 not coming into contact

Table 1. Factors analyzed for the optimization of the degradation of biodiesel

Experimenta [Fe2+] / (mol L-1) [H2O2] / (mol L-1) Fenton volume / mL [Fe²+] : [H2O2] Light
1 5.34 × 10-3 8.05 × 10-3 1.54 0.65 absence
2 10.4 × 10-3 32.9 × 10-3 3.00 0.30 presence
3 5.34 × 10-3 4.02 × 10-3 1.54 1.33 presence
4 10.4 × 10 -3
15.6 × 10 -3
3.00 0.63 absence
5 5.34 × 10-3 8.05 × 10-3 1.54 0.65 presence
6 5.34 × 10-3 4.02 × 10-3 1.54 1.33 absence
7 10.4 × 10 -3
32.9 × 10 -3
3.00 0.30 absence
8 10.4 × 10-3 15.6 × 10-3 3.00 0.63 presence
Each experiment was repeated five times.
a
Vol. 25, No. 4, 2014 Ambrosio et al. 763

Table 2. GC/FID and COD results obtained for the degradation of biodiesela

FAMEb Total FAME / CODc /

Sample
16:0 / (mg g )
-1
18:0 / (mg g )
-1
18:1 / (mg g )
-1
18:2 / (mg g )
-1
18:3 / (mg g )
-1 (mg g-1) (g L-1)

B 100 150.58 ± 20.585 86.72 ± 7.337 255.81 ± 19.824 309.77 ± 24.112 32.85 ± 2.741 835.73 ± 9.307 –
Exp. 1 161.64 ± 6.214 96.37 ± 1.034 281.78 ± 9.651 337.84 ± 10.618 36.45± 1.082 914.08 ± 4.559 1.197 ± 0.639
Exp. 2 353.03 ± 34.363 210.55 ± 22.019 69.60 ± 6.909 4.10 ± 2.064 NI 637.28 ± 14.718 21.671 ± 2.006
Exp. 3 172.69 ± 58.105 104.00 ± 36.886 26.36 ± 8.658 NI NI 303.05 ± 24.806 20.671 ± 0.550
Exp. 4 154.03 ± 6.331 92.07 ± 3.439 273.04 ± 7.566 318.51 ± 9.934 33.78 ± 1.394 871.43 ± 3.371 1.312 ± 0.423
Exp. 5 140.91 ± 0.741 86.03 ± 2.514 23.40 ± 0.703 0.57 ± 0.589 NI 250.91 ± 1.962 23.353 ± 2.611
Exp. 6 125.07 ± 3.809 70.36 ± 4.358 221.17 ± 6.863 253.58 ± 8.118 26.38 ± 1.579 696.55 ± 2.584 5.024 ± 4.622
Exp. 7 160.51 ± 5.046 94.29 ± 4.429 291.10 ± 11.990 342.25 ± 7.915 35.85 ± 1.458 924.00 ± 3.982 1.233 ± 0.175
Exp. 8 225.85 ± 84.277 138.01 ± 50.769 44.65 ± 16.058 NI NI 408.51 ± 34.111 24.173 ± 1.240
Photolysis 162.08 ± 26.749 100.95 ± 17.344 114.69 ± 20.713 NI NI 377.71 ± 4.765 74.800 ± 17.403
a
Results expressed in mg of ester for each gram of biodiesel or organic phase analyzed. Values are means of three replicates accompanied by deviation;
b
the symbols represent the main chain of the FAME (fatty acid methyl ester); cchemical oxygen demand: expressed in grams of O2 consumed per liter of
sample water phase analyzed. Values are means of five repetitions accompanied by its standard deviation; NI: not identified; Exp.: experiments which were
carried out varying the analysis conditions, as described in Table 1.

with water prior to the GC/FID analysis. When B 100
contacted water, a fraction of glycerol migrated from
B 100 to the aqueous phase, causing a relative increase
in the concentration of FAMEs in the organic phase. This
effect occurred in all experiments but was most evident
in experiments 1, 4, and 7, which were performed in the
absence of light.
During the degradation, there was a decrease in
percentage in 16:1, 18:1, 18:2, and 18:3 unsaturated esters
and an increase in the proportion of 16:0, 17:0, 18:0, 20:0,
22:0, and 24:0 saturated esters. Photolysis was performed
to determine the influence of light alone on this process,
proving to be an intense factor, as there was a decrease in
total FAMEs of approximately 55% in relation to their initial
amount in B 100. In contrast, a high COD, 59.476 g L-1, was
observed for the aqueous phase. In the assays with Fenton’s
reagent, the maximum COD was lower than half of this
value, which can be attributed to the action of the reagent
on the photoproducts that migrated to the aqueous phase,
thus accelerating their degradation. In the experiments in
which the Petri dishes were wrapped with aluminum foil
(experiments 1, 4, 6, and 7), the only effects observed
were those of Fenton’s reagent and the temperature. As the
increase in the temperature inside the irradiation box was
not greater than 2 °C above room temperature, the main
effect observed was due to Fenton’s reagent in the absence
of light. Additionally, the COD values of the aqueous phase
were approximately 20% smaller than those observed for
the aqueous phase under irradiation, suggesting that few
organic-phase components were transported to the aqueous
phase. Therefore, the joint action of Fenton’s reagent and
light is important in these reactions.
The aliquots with Fenton’s reagent that were irradiated
showed an increase in saturated FAMEs and a decrease in
unsaturated FAMEs, with a mean value of 61.5 ± 2.05%
relative to B 100 and 22.5 ± 1.59% for the aqueous-phase
COD. The best results were obtained in experiment 5, which
had a total FAME content of 229.54 mg per g of organic
phase analyzed and a smaller aqueous phase COD of 69%
relative to the photolysis assay.
Gas chromatography coupled with mass spectrometry
(GC/MS)
Kovats experimental indices (KI) were calculated using
equation 5, as proposed by Van Den Dool and Kratz:15
KI = 100 × n + 100 Δn × [(trx – trz) / (trz+1 – trz)] (5)
where n = the number of carbon atoms of compound x,
Δn = the variation in the number of carbon atoms of the
hydrocarbon standard eluted before and after the addition of
compound x, trx = the compound x retention time, trz = the
retention time of the hydrocarbon standard eluted before
the addition of compound x, and trz+1 = the retention time
of the hydrocarbon standard eluted immediately after the
addition of compound x.
The results were compared with those in the literature
(NIST MS search, 2.0) and are provided in Table 3.
B 100 presented four main FAMEs: 16:0, 18:2, 18:1,
and 18:0, and two other compounds, 16:1 and nonanoic
acid 9-oxo-methyl ester, were found during the 24 h
photodegradation process. In addition to these compounds,
ramifications of the ester carbon chain and epoxidation were
764 Study of Biodiesel Photodegradation Through Reactions Catalyzed by Fenton’s Reagent J. Braz. Chem. Soc.

Table 3. GC/MS results and Kovats Index (KI) of the biodiesel analyzed during photodegradation with 1.54 mL of Fenton’s reagent, [Fe²+]:[H2O2] = 0.65,
for up to 360 h

Sample Retention time / min No. of C Experimental KI Compound Theoretical KI

Standard 26.97 14 – n-tetradecane 1400
31.03 15 – n-pentadecane 1500
38.58 17 – n-heptadecane 1700
42.09 18 – n-octadecane 1800
48.62 20 – n-eicosane 2000
60.13 24 – n-tetracosane 2400
B 100 46.13 17 1924 hexadecanoic acid, methyl ester 1870-1926
51.36 19 2095 9,12-octadecadienoic acid (Z,Z)-, methyl ester 2092-2098
51.55 19 2102 9-octadecenoic acid, methyl ester 2082-2107
52.33 19 2129 octadecanoic acid, methyl ester 2128-2135
24 h 28.20 10 1430 nonanoic acid, 9-oxo-, methyl ester 1436-1439
46.10 17 1923 hexadecanoic acid, methyl ester 1870-1926
51.31 19 2093 9,12-octadecadienoic acid (Z,Z)-, methyl ester 2092-2098
51.50 19 2100 9-octadecenoic acid (Z,Z)-, methyl ester 2082-2107
51.68 19 2106 10-octadecenoic acid, methyl ester 2100-2110
52.32 19 2129 octadecanoic acid, methyl ester 2128-2135
96 h 28.19 10 1430 nonanoic acid, 9-oxo-, methyl ester 1436-1439
46.13 17 1924 hexadecanoic acid, methyl ester 1870-1926
51.33 19 2094 9,12-octadecadienoic acid (Z,Z)-, methyl ester 2092-2098
51.52 19 2101 9-octadecenoic acid (Z,Z)-, methyl ester 2082-2107
51.72 19 2108 10-octadecenoic acid, methyl ester 2100-2110
52.32 19 2129 octadecanoic acid, methyl ester 2128-2135
56.68 19 2280 octadecanoic acid, 9,10-epoxy-methyl ester 2129
58.71 18 2351 6-hexadecenoic acid, 7-methyl, methyl ester 1963
168 h 28.23 10 1431 nonanoic acid, 9-oxo-, methyl ester 1436-1439
46.10 17 1923 hexadecanoic acid, methyl ester 1870-1926
51.52 19 2101 10-octadecenoic acid, methyl ester 2100-2110
52.33 19 2129 octadecanoic acid, methyl ester 2128-2135
56.69 19 2070 octadecanoic acid, 9,10-epoxy-methyl ester 2129
58.71 57 2088 6-hexadecenoic acid, 7-methyl,methyl ester 1963
360 h 36.98 15 1706 tetradecanoic acid, methyl ester 1675-1725
45.58 17 1907 Hexadecanoic acid, methyl ester 1870-1926
50.97 19 2082 10-octadecenoic acid, methyl ester 2100-2110
51.80 19 2111 octadecanoic acid, methyl ester 2128-2135
56.15 19 2262 octadecanoic acid, 9,10-epoxy-methyl ester 2129

also observed after 96 h; however, the experimental KI values
did not match the theoretical values. No KI values were found
in the literature for the analysis of this compound using the
column used in the present study. Ketones and epoxides have
also been found during the thermal degradation of biodiesel,5
and it was observed that saturated FAMEs were not oxidized
at high temperatures, with high concentrations remaining
after 360 h. Ketones were observed during the degradation
of epoxides, which are produced by the addition of peracids
to carbon chain unsaturations. During photodegradation,
there was an increase in the amount of saturated esters and
a decrease in the amount of unsaturated esters. However, the
GC/MS analysis was only qualitative; thus, a quantitative
analysis was performed using GC/FID.
Vol. 25, No. 4, 2014 Ambrosio et al. 765

Gas chromatography coupled with flame ionization detection ¹H NMR

(CG/FID)
The ¹H NMR spectra exhibited signals for specific
The FAMEs initially found in B100 were 16:0, 17:0, types of protons in the ester mixture, as illustrated by the
18:0, 20:0, 22:0, 24:0, 16:1, 18:1, 18:2, and 18:3, with the spectrum of B 100 shown in Figure 1.
main esters being quantified during the photodegradation Some researchers have used ¹H NMR spectra to analyze
process after 144 and 360 h. The results are given in Table 4. fatty acid and biodiesel composition by utilizing only the
integration value peaks caused by the protons in the fatty
Table 4. FAMEs quantified by GC/FID for pure biodiesel (B 100) and
biodiesel degraded by 1.54 mL of Fenton’s reagent, [Fe²+]:[H2O2] = 0.65, acid chains.3,5,18 The peaks that may allow the quantification
for 144 h and 360 ha of unsaturated FAMEs are those of the protons of olefins
(5.3-5.4 ppm), bis-allylic carbons (2.7-2.8 ppm), allylic
FAME B 100 / (mg g-1) 144 h / (mg g-1) 360 h / (mg g-1) carbons (2.0-2.1 ppm), and terminal methyl groups
16:0 b
150.58 ± 20.585 225.85 ± 84.277
a b
144.15 ± 3.005 (0.8‑0.9 ppm). Saturated FAMEs can be determined using
18:0 b
86.72 ± 7.337 138.01 ± 50.769
a b
84.98 ± 2.520 the signal for methylene protons, CH2 (1.2-1.4 ppm). The
18:1 255.81 ± 19.824
a b
44.65 ± 16.058 – % FAME values calculated according to Knothe et al.3,18
18:2 309.77 ± 24.112
a
– – are given in Table 5.
18:3 a
32.85 ± 2.741 – –
Table 5. FAME percentages of non-irradiated B 100 and B 100 irradiated
Total 835.73 ± 38.203
a b
408.51 ± 99.689 229.13 ± 3.922
c
for 12 h by 1.54 mL of Fenton’s reagent, [Fe²+]:[H2O2] = 0.65, analyzed
a
The results are averages of three replicates followed by the standard by 1H NMR. The samples were dissolved in CDCl3
deviation. Expressed as mg FAME per 1 g of organic material analyzed.
Means followed by the same letter on the same line do not differ FAMEs / %
significantly (Tukey’s test, p < 0.05). Sample
C16:0 C18:0 C18:1 C18:2 C18:3
B 100 7.59 46.25 31.36 9.35 5.45
FAMEs, particularly the unsaturated FAMEs, were
12 h 13.65 58.23 19.12 5.78 3.22
degraded during the photodegradation process. A 1 g
sample of B 100 showed 835.73 mg FAMEs. In contrast,
there was a total decrease in FAMEs of 51% after 144 h, The peaks of the terminal methyl groups of linolenic
and only 27% of the initial amount was observed after methyl esters were not observed after 24 h of irradiation,
360 h. Furthermore, only 16:0 and 18:0 were detected after suggesting that the conjugated double bonds of FAMEs
degradation for 360 h. were oxidized during this period. This result is due to the

Figure 1. ¹H NMR spectrum of B 100. The biodiesel was dissolved in CDCl3 at 0.1% v/v TMS (Aldrich, δ = 0.0 ppm).
766 Study of Biodiesel Photodegradation Through Reactions Catalyzed by Fenton’s Reagent
Figure 2. ¹H NMR spectra of non-irradiated B 100 and B 100 irradiated with 1.54 mL of Fenton’s reagent, [Fe²+]:[H2O2] = 0.65, for up to 120 h. The
samples were dissolved in CDCl3 at 0.1% v/v TMS.
proximity of the C15-C16 double bond in linolenic methyl
esters to the terminal CH3; the signal of the terminal CH3
is thus shifted downfield to approximately 0.95 ppm and
can be integrated separately from the signal of the terminal
CH3 in the other fatty acid chains.18 As peaks for olefinic,
allylic, and bis-allylic carbon protons were not observed
after 120 h of irradiation, the fatty acid unsaturations had
been oxidized during this time period (Figure 2).
An increase in viscosity and darkening of the color
of the degraded biodiesel was observed, most likely
due to the oxidation of unsaturations. The signal of the
beta group (–CH2–CH2–COOCH3) at δ 1.6 ppm showed
some broadening during the photodegradation process,
which may be due to the oxidation of unsaturations and
consequent overlapping of CH2 group signals.
Oxitest
The Oxitest equipment analyzes the induction time
of organic matter, which is the time needed to oxidize
organic matter in an oxygen-rich atmosphere, a process
that depends on the nature and amount of fatty acids
in the biodiesel sample. Biodiesel free of antioxidants
has an induction time of approximately 2 h,25 and the
investigated biodiesel was considered to be free of
antioxidants because it showed an induction time of
2:44 h. After irradiation of the samples with Fenton’s
J. Braz. Chem. Soc.
reagent for 6 h, the induction time decreased to 29 min,
and the samples were completely oxidized in under 24 h
of photodegradation.
Chemical oxygen demand (COD)
The aqueous-phase COD increased during up to
120 h of photodegradation with 3.0 mL of Fenton’s
reagent ([Fe2+]:[H2O2] = 0.65, with 10.4 × 10-3 mol L-1
[Fe2+] and 15.6 × 10-3 mol L-1 [H2O2]) in different Petri
dishes (Figure 3). This increase may be attributed to an
increase in the amount of organic matter in the water
from photoproducts transferred from the organic phase
to the aqueous phase. After this time, the COD decreased
progressively up to 360 h of photodegradation and
apparently reached equilibrium, with the migration of the
photoproducts to the aqueous phase, following oxidation
by Fenton’s reagent in the presence of light. However, the
COD was not completely reduced because the organic
matter resulting from the degradation of biodiesel continues
to migrate to the aqueous phase until the biodiesel film on
the water surface has been completely degraded.
Toxicity test with Artemia salina L.
There was an increase in the mortality rate of
A. salina L. during photodegradation, reaching a maximum
Vol. 25, No. 4, 2014
Figure 3. Chemical oxygen demand of the aqueous phase under irradiation
for 360 h with 3.0 mL of Fenton’s reagent: [Fe2+]:[H2O2] = 0.65.
between 120 and 168 h, though we observed a slight
decrease in mortality at 360 h for the most diluted aliquots
(1%). The aqueous-phase samples with a concentration
greater than 20% presented 100% mortality for all the
photodegradation times investigated, indicating that the
biodiesel products were either transferred or transformed
during photodegradation. The experimental results of the
A. salina L. assays are provided in Figure 4.
Figure 4. Percent mortality rate of A. salina L. in aliquots of aqueous
phases irradiated for 360 h with 1.54 mL of Fenton’s reagent,
[Fe²+]:[H2O2] = 0.65. The aqueous phase was diluted with a saline mixture
(v/v): (--) 1%; (--) 10%; (--) 20%.
By comparing the results of the acute toxicity results
of COD (Figure 3), it is clear that recalcitrant components
were formed, reaching higher concentrations between
120 h and 168 h of photodegradation. After 168 h, there
was a decrease in these recalcitrant compounds due to the
action of the photo-Fenton system, thereby decreasing the
mortality of the microcrustaceans.
Biodegradation refers to the process by which
organisms use organic contaminants as a food or energy
source. However, recalcitrant compounds are not degraded
by microorganisms; thus, the action of a chemical agent is
very important, as noted in the work of Garcia et al.22 in
Ambrosio et al. 767
which the action of advanced oxidation processes (AOP)
decreased the toxicity of an effluent to A. salina L.
A study of the biodegradability of biodiesel in soil
found 84% biodegradation of B 100 after 60 days.26 This
same study also reported respirometer tests and the analysis
of microorganism DNA, concluding that, although B 100
can be more rapidly degraded than diesel fuel, there was a
decrease in the microbial community of soil contaminated
with B 100 after 60 days. This finding indicates that some
products of biodiesel photodegradation can be toxic to
organisms.
Because there was no mortality with the aqueous phase
in contact with biodiesel not degraded after 24 h, the
toxic effects of photodegradation can be attributed to the
biodiesel intermediates that migrated to the aqueous phase.
Fluorescence spectroscopy
Synchronous scan excitation spectra were obtained by
measuring the fluorescence intensity simultaneously with
excitation and emission at a certain wavelength using a
constant optimized by the difference of the wavelengths,
as follows: Δλ = λ em – λ exc. 27 The synchronous scan
excitation spectra of the undiluted organic phase are shown
in Figure 5.
Figure 5. Synchronous fluorescence spectra of (a) biodiesel B 100 and
biodiesel with 1.54 mL of Fenton’s reagent, [Fe²+]:[H2O2] = 0.65, irradiated
for (b) 24 h; (c) 72 h; and (d) 120 h.
Prior to photodegradation, the biodiesel presented bands
between 340 and 363 nm and close to 520 nm. However,
only one band, between 380 and 400 nm, was observed after
24 h of photodegradation, and this band became less intense
at 72 h; no fluorescence signal was observed at 120 h. The
chromophore compounds that may be present in biodiesel,
such as FAME molecules with double conjugated bonds, are
oxidized during the photodegradation process, eliminating
768 Study of Biodiesel Photodegradation Through Reactions Catalyzed by Fenton’s Reagent
the fluorescence. Although the aqueous-phase aliquots were
scanned, the salts previously added to interrupt the Fenton
reaction interfered with the measurements.
UV/Vis photometry
During the photodegradation process, the biodiesel
photoproducts are transferred to water, which increases
absorption in the UV region. As the aqueous phase had
a maximum absorbance at 226 nm, the absorbance was
monitored at this wavelength for various photodegradation
times, as shown in Figure 6. The absorbance was at a
maximum after 120 h of irradiation, decreasing thereafter
and reaching a minimum at 360 h, most likely because
the photoproducts were transferred to water and continued
to be subjected to the action of Fenton’s reagent and
irradiation. Two maximum absorbance wavelengths
were observed for biodiesel in the UV region, at 205
and 228 nm. During irradiation, the absorbance at 205
nm decreased and reached a minimum after 360 h of
photodegradation. An increase in absorbance was also
observed at 228 nm during the first 24 h, followed by a
decrease in this region up to 360 h.
Figure 6. UV-region spectrophotometry of the aqueous phase (- -)
λ = 226 nm (aqueous phase: distilled water dilution rate, 1:1000).
UV-region spectrophotometry of the organic phase (--) λ = 205 and
(--) λ = 228 nm (organic phase: ethanol dilution rate, 1:1000). For
degradation of 1 mL of biodiesel, 1.54 mL of Fenton’s reagent were used,
[Fe²+]:[H2O2] = 0.65.
Conclusions
The best degradation conditions for 1 mL of biodiesel
using Fenton’s reagent were [Fe²+]:[H2O2] = 0.65, with the
addition of 1.54 mL of a mixture of sulfate heptahydrate
with concentrations of [Fe2+] = 5.34 × 10-3 mol L-1 and
[H2O2] = 8.05 × 10-3 mol L-1 under irradiation. Under these
J. Braz. Chem. Soc.
conditions, the FAME concentrations were the smallest in
the organic phase, and COD was smaller in the aqueous
phase than in the non-irradiated samples and the samples
irradiated without the addition of Fenton’s reagent. After
360 h, a 73% decrease in total FAMEs was found, with
only 18:0 and 16:0, the most persistent FAMEs, being
quantified by GC/FID. Peaks for ketone and carbon chain
epoxide groups were observed during photodegradation
according to GC/MS analyses, as were typical aldehyde
and short-chain fatty acid shifts according to 1H NMR
analyses. Ecotoxicity analyses with Artemia salina revealed
toxic components in the aqueous phase, which increased
up to 168 h of photodegradation and decreased thereafter.
Supplementary Information
Supplementary data are available free of charge at
http://jbcs.sbq.org.br as PDF file.
Acknowledgements
The authors would like to thank Coordenação de
Aperfeiçoamento de Pessoal de Nível Superior (CAPES)
and Conselho Nacional de Desenvolvimento Científico e
Tecnológico (CNPq) and Fundação Araucária for research
funds.
References
1. http://www.biodieselbr.com/noticias/usinas/acidente/caminhao-
biodiesel-tomba-contamina-lago-palmas-190111.htm accessed
in January 2014.
2. Dunn, R. O.; J. Am. Oil. Chem. Soc. 2002, 79, 915.
3. Knothe, G.; Eur. J. Lipid. Sci. Technol. 2006, 108, 493.
4. Knothe, G.; Fuel Process. Technol. 2007, 88, 669.
5. Chuck, C. J.; Bannister, C. D.; Jenkins, R. W.; Lowe, J. P.;
Davidson, M. G.; Fuel 2012, 96, 426.
6. Aquino, I. P.; Hernandez, R. P. B.; Chicoma, D. L.; Pinto,
H. P. F.; Aoki, I. V.; Fuel 2012, 102, 795.
7. Sarika, R.; Kalogerakis, N.; Mantzavinos, D.; Environ. Int.
2005, 31, 297.
8. Ginos, A.; Manios, T.; Mantzavinos, D.; J. Hazard. Mater. 2006,
B133, 135.
9. Filidei, S.; Masciandaro, G.; Ceccanti, B.; Water Air Soil Pollut.
2003, 145, 79.
10. Sabbah, I.; Marsook, T.; Basheer, S.; Process Biochem. 2004,
39, 1947.
11. Ramírez, X. M. V.; Mejía, G. M. H.; López, K. V. P.; Vásquez,
G. R.; Sepúlveda, J. M. M.; Water Sci. Technol. 2012, 66, 824.
12. Nogueira, R. F. P.; Trovó, A. G.; Silva, M. R. A.; Villa, R. D.;
Quim. Nova 2007, 30, 400.
Vol. 25, No. 4, 2014
13. Galvão, S. A. O.; Mota, A. L. N.; Silva, D. N.; Moraes, J. E. F.;
Sci. Total Environ. 2006, 367, 42.
14. Silva, M. R. A.; Oliveira, M. C.; Nogueira, R. F. P.; Eclet. Quim.
2004, 29, 19.
15. Dool, H. V. D.; Kratz, D. J.; J. Chromatogr. 1963, 11, 463.
16. Joseph, J. D.; Ackman, R. G.; J. AOAC Int. 1992, 75, 488.
17. Visentainer, J. V.; Quim. Nova 2012, 35, 274.
18. Knothe, G.; Kenar, J. A.; Eur. J. Lipid Sci. Technol. 2004, 106,
88.
19. Greenberg, A. E.; Clesceri, L. S.; Eaton, A. D. In Standard
Methods for the Examination of Water and Wastewater, 8 ed.; th
Greenberg, A. E.; Clesceri, L. S.; Eaton, A. D., eds.; APHA:
Washington, D. C., 1992, ch. 5.1.
20. Nogueira, R. F. P.; Oliveira, M. C.; Paterlini, W. C.; Talanta
2005, 66, 86.
21. Meyer, B. N.; Ferrigini, N. R.; Putnan, J. E.; Jacobsen, L. B.;
Nichols, D. E.; McLaughlin, J. L.; Planta Med. 1982, 45, 31.
Ambrosio et al. 769
22. Garcia, J. C.; Freitas, T. K. F. S.; Palácio, S. M.; Ambrosio, E.;
Souza, M. T. F.; Santos, L. B.; Almeida, V. C.; de Souza, N. E.;
Environ. Monit. Assess. 2013, 185, 2179.
23. D’Auria, M.; Emanuele, L.; Racioppi, R.; Velluzzi, V.;
J. Photochem. Photobiol. A 2008, 198, 156.
24. Gracetto, A. C.; Batistela, V. R.; Caetano, W.; de Oliveira,
H. P. M.; Santos, W. G.; Cavalheiro, C. C. S.; Hioka, N.; J. Braz.
Chem. Soc. 2010, 21, 1497.
25. Araújo, F. D. S.; de Moura, C. V. R.; Chaves, M. H.; Quim.
Nova 2010, 33, 1671.
26. Silva, G. S.; Marques, E. L. S.; Dias, J. C. T.; Lobo, I. P.;
Gross, E.; Brendel, M.; da Cruz, R. S.; Rezende, R. P.; Appl.
Soil Ecol. 2012, 55, 27.
27. Miano, T. M.; Senesi, N.; Sci. Total Environ. 1992, 117/118,
41.
Submitted: October 10, 2013
Published online: February 11, 2014
 SI
J. Braz. Chem. Soc., Vol. 25, No. 4, S1-S5, 2014.

Supplementary Information
Printed in Brazil - ©2014 Sociedade Brasileira de Química
0103 - 5053 $6.00+0.00

Study of Biodiesel Photodegradation Through Reactions

Catalyzed by Fenton’s Reagent

Elizangela Ambrosio, Lorena M. Milano, Maísa Tatiane F. de Souza,

Lucas U. R. Chiavelli, Paula F. Montanher, Jesuí V. Visentainer,
Vitor de C. Almeida, Nilson E. Souza* and Juliana C. Garcia

Departamento de Química, Universidade Estadual de Maringá,

Av. Colombo, 5790, 87020-900 Maringá-PR, Brazil

The absorbance spectrum for ethanol compared with the

absorbance spectrum of biodiesel is shown in Figure S1.

Figure S1. UV-region spectrophotometry of the biodiesel (ethanol dilution

rate, 1:1000) (); ethanol (99.3%, FMaia) (); and baseline ().

It is possible to use ethanol as solvent for the organic

phase as it does not interfere with their absorbance spectra.
Figure S2 shows the absorption spectra for the organic
phase and the aqueous phase which have undergone the
action of Fenton’s reagent and light at various periods of
time.

Figure S2. Absorbance spectra of: a) organic phase and b) the aqueous
phase irradiated for 360 h using Fenton’s reagent: () distilled water
(baseline), () ethanol (baseline), () biodiesel undegraded, () 24 h,
() 72 h, () 120 h, () 168 h, () 360 h. Aqueous phase was diluted
with distilled water, 1:1000 and the organic layer diluted with ethanol,
1:10000. For degradation of 1 mL biodiesel, Fenton’s reagent was used
[Fe²+]:[H2O2] = 0.65, 1.54 mL was added to a mixture of ferrous sulfate
heptahydrate concentration of [Fe2+] = 5.34 × 10-3 mol L-1 and [H2O2] =
8.05 × 10-3 mol L-1 in the presence of light.
*e-mail: nesouza@uem.br
S2 Study of Biodiesel Photodegradation Through Reactions Catalyzed by Fenton’s Reagent J. Braz. Chem. Soc.

Figure S3. GC/MS chromatogram of B 100 dissolved in dichloromethane.

Figure S4. GC/MS chromatogram of degraded biodiesel by photo-Fenton ([Fe²+]:[H2O2] = 0.65, 1.54 mL) for 24 h, dissolved in dichloromethane.
Vol. 25, No. 4, 2014 Ambrosio et al. S3

Figure S5. GC/MS chromatogram of degraded biodiesel by photo-Fenton ([Fe²+]:[H2O2] = 0.65, 1.54 mL) for 96 h, dissolved in dichloromethane.

Figure S6. GC/MS chromatogram of degraded biodiesel by photo-Fenton ([Fe²+]:[H2O2] = 0.65, 1.54 mL) for 168 h, dissolved in dichloromethane.
S4 Study of Biodiesel Photodegradation Through Reactions Catalyzed by Fenton’s Reagent J. Braz. Chem. Soc.

Figure S7. GC/MS chromatogram of degraded biodiesel by photo-Fenton ([Fe²+]:[H2O2] = 0.65, 1.54 mL) for 360 h, dissolved in dichloromethane.

Figure S8. GC/MS chromatogram of a mixture of alkane standards (C9‑C21) dissolved in dichloromethane.
Vol. 25, No. 4, 2014 Ambrosio et al. S5

Figure S9. ¹H NMR spectrum of degraded biodiesel by photo-Fenton ([Fe²+]:[H2O2] = 0.65, 1.54 mL) for 360 h. Typical shifts can be observed for aldehyde
(9.4‑10.0 ppm) and carboxylic acids (about 8 ppm), whereas the chemical shifts between 6.0 and 6.4 ppm are due to the conjugated bonds of these compounds.